Research On Energy Consumption And Correlation Performance Optimization Of Mixed-criticality Real-time System

Real-time embedded system has been widely applied to various fields along with the rapid development of computer technology.The embedded system evolved from a previous single function to an increasingly complex scale of functionality,which led to the increasing cost of embedded systems.In order to save costs,there has been a tendency to integrate the different key functional applications deployed in a standalone subsystem into a common hardware platform that shares processor resources,resulting in the emergence of mixed-critical systems.In the mixed-critical system,the failure of the execution of certain tasks can cause catastrophic consequences,and the failure of some tasks will only bring down the user experience.So we need to provide different levels of protection for different critical levels of application in mixed-critical systems.The rapid development of embedded device performance has greatly expanded its application range,and the embedded device battery capacity and volume constraints seriously hampered the performance of embedded devices.Therefore,abstracting the specific application model according to the specific application scenarios and establishing the constraints of high critical level task implementation and embedded system real-time and low energy consumption.It is very meaningful to find the appropriate low-power scheduling strategy under the premise of ensuring the correct execution of the high-critical task of the mixed-critical system and the real-time performance of the embedded system.This thesis focuses on the optimal design of mixed-critical real-time system for task scheduling from the aspects of high critical task security,system real-time and energy consumption.Based on the static task scheduling,this thesis optimizes the energy consumption and related performance of the mixed-critical real-time system with considering the constraints of the system.Firstly,we choose the application scenario of UAV flight control and establish a safety-sensitive mixed-critical system model.And we optimize the system energy consumption the condition that all system constrains are guaranteed.We have also established a reliability-sensitive mixed-critical system model,and optimize the system safety the condition that all system constrains are satisfied.We discuss distributed mixed-critical real-time embedded applications,and we establish the mixed-critical distributed system model for the applications.A scheduling strategy for distributed mixed-critical real-time system was proposed to optimize the energy consumption of distributed mixed-critical real-time systems.